Twin concordances test for ascertained trichotomous traits data.

Area of Science:

• Human genetics
• Biostatistics
• Twin studies

Background:

• Twin studies investigate genetic influence on diseases using various similarity measures for dichotomous traits.
• Ascertainment procedures in twin studies create truncated data, invalidating odds ratios and correlation estimations.
• Existing concordance measures for dichotomous traits can be estimated via the likelihood method, but extensions to trichotomous traits are underexplored.

Purpose of the Study:

• To address the lack of defined concordance measures and estimation methods for ascertained trichotomous traits data.
• To introduce novel casewise and pairwise concordance measures specifically for trichotomous traits.
• To develop and validate a practical estimation method for twin concordances in ascertained trichotomous trait data.

Main Methods:

• Definition of new casewise and pairwise concordance measures for trichotomous traits.
• Application of the self-contained subsets method (SCSM) for estimating twin concordances in ascertained data.
• Derivation of asymptotic variances for SCSM estimates and establishment of a testing procedure for comparing monozygotic and dizygotic twin concordances.

Main Results:

• The SCSM provides estimates equivalent to the likelihood method but with greater ease of application.
• Asymptotic variances for SCSM estimates under ascertainment were derived.
• A method for testing the equality of concordance measures between monozygotic and dizygotic twin pairs was established.

Conclusions:

• The proposed concordance measures and SCSM offer a viable approach for analyzing ascertained trichotomous traits in twin studies.
• The SCSM simplifies the estimation process compared to traditional likelihood methods.
• The developed methods facilitate the investigation of genetic influences on complex traits using twin data.